AAA+ proteases are responsible for intracellular protein degradation in all domains of life and controlled ATP-dependent protein degradation and regulated gene expression are partners in defining the proteome. Furthermore, the same enzymatic machinery used for degradation is critical for disassembling protein complexes and antagonizing protein aggregation. As both degradation and disassembly are inherently destructive processes and the efficiency of substrate breakdown is often made at the level of substrate choice, it is imperative to understand the molecular principles guiding substrate selection by these ATP-driven, protein-destruction machines. This proposal focuses on three areas of substrate selection of the bacterial AAA+ proteases ClpXP, ClpAP and Lon. Aim 1 concentrates on elucidating the design principles and molecular contacts used by two specific classes of ClpX substrate-recognition signals. As part of this aim, we test the hyphothesis that some signals are designed such that they are preferentially recognized in the context of multi-protein complexes. We also propose to solve the structure of newly-discovered ClpX N-domain-interacting motifs bound to the N-domain to further understand the mechanistic basis of ClpX recognition. Aim 2 dissects how Lon and ClpAP proteases and the E. coli sHsps (IbpA and IbpB) interact, tests models for how these interactions impact protein quality-control pathways. We are excited to explore the functional consequences of this newly-discovered intersection between the aggregation-prevention (sHsp) and protein-destruction (protease) arms of the protein quality-control network. The final aim begins to address mechanisms for control of protein degradation in response to oxidative stress and oxidative damage. Two proteins are chosen for initial studies: the B. subtilis peroxide-sensing transcription factor, PerR and the E. coli mini-ferritin, Dps. Preliminary data indicate that PerR is specifically subject to accelerated degradation by LonA protease when its histidine active center is irreversibly oxidized. In contrast to PerR, Dps degradation is inhibited by peroxide. We propose to isolate factors responsible for these examples of environmentally-controlled regulation. Successful completion of these aims will provide substainal new insights into protease recognition and may uncover new paradigms for contol of the proteome.